Hydroxymethyl-indolizidines and quinolizidines and their use as α-glucosidase I inhibitors

A number of hydroxylated derivatives of indolizidine and pyrrolizidine have been reported in the literature. For the most part, these compounds have been isolated from natural sources and the best known compound of this type is castanospermine which can also be named as [1S(α,6β,7α,8β, 8aβ)]-octahydro-1,6,7,8-indolizinetetrol. Pyrrolizidines having a carbon substituent (i.e., a hydroxymethyl substituent) have been reported previously [see Nash et al., Tet. Letters, 29(20), 2487 (1988); Nash et al., Tetrahedron, 44, 5959(1988)]but similarly substituted indolizidine and quinolizidines do not appear to have been described previously.

The present invention is directed to hydroxylated indolizidines and quinolizidines having a hydroxymethyl substituent and to esters of such compounds. More particularly, the present invention is directed to compounds having the following general formulas: ##STR1## wherein n is 1 or 2; R 1 and R 2 are independently H or OA4; A, A 1 , A 2 , A 3 and A 4 are each independently hydrogen, Cl-18 alkanoyl or ##STR2## and Y, Y', and Y" are each independently hydrogen, C 1-4 alkyl, C 1-4 alkoxy rr halogen; or the pharmaceutically acceptable salts of the aforesaid compounds.

In the above structural formula, the wavy line indicates that the second ring can be fused to the six-membered (piperidine) ring shown in either of the two isomeric forms possible. In addition, where the groups R 1 and R 2 represent OA 4 , it should be recognized that there is a secon attached at the same ring carbon and that second atom is hydrogen. Furthermore, the groups R 1 and R 2 can have either of the two possible configurations with respect to the ring structure.

The C 1-18 alkanoyl grOupS referred to above can be straight- or branched-chain and can be exemplified by formyl, acetyl, propionyl, butyryl, isobutyryl, hexanoyl, octanoyl, decanoyl and hexadecanoyl. The halogens referred to above can be exemplified by fluorine, chlorine, bromine or iodine. The C 1-4 alkyl groups referred to above, whether alone or part of an alkoxy group, can be straight- or branched-chain alkyl groups containing up to 4 carbon atoms. Examples of various such groups are methyl, ethyl, propyl, butyl, methoxy, ethoxy or butoxy.

Acid addition salts with pharmaceutically acceptable acids referred to above are equivalent to the amines for the purposes of this invention. Illustrative of such salts are the salts with inor9anic acids such as, for example, hydrochloric, hydrobromic, sulfuric, phosphoric and like acids; with organic carboxylic acids such as, for example, acetic, propionic, glycolic, lactic, pyruvic, malonic, succinic, fumaric, malic, tartaric, citric, ascorbic, maleic, hydroxymaleic and dihydroxymaleic, benzoic, phenylacetic, 4-amino-benzoic, 4-hydroxybenzoic, anthranilic, cinnamic, salicylic, 4-aminosalicylic, 2-phenoxybenzoic, 2-acetoxybenzoic, mandelic and like acids; and with organic sulfonic acids such as methanesulfonic acid and p-toluenesulfonic acid. Such salts can be obtained by standard procedures from an amine of this invention and the appropriate acid.

In the preparation of the indolizidine compounds of the present invention wherein R 1 and R 2 are both hydrogen or --OH, 2,6-dideoxy-2,6-[[(phenylmethoxy)carbonyl]imino]-1,3,4,5-tetrakis -O-(phenylmethyl)-D-glycero-L-gulo-heptitol serves as the basic starting material. The specific series of reactions used to convert this compound to a hydroxylated indolizidine is shown in Scheme A below. In the structural formulas, Bn is phenylmethyl and Z is benzyloxycarbonyl or t-butyloxycarbonyl. ##STR3##

Thus, the free hydroxymethyl group in the starting compound is oxidized to the corresponding aldehyde (II). Oxalyl chloride and dimethylsulfoxide in methylene chloride (Swern oxidation) can be used for this reaction. Base catalysed epimerization with 1,8-diazabicyclo[5.4.0]undec-7-ene or 1,4-diazabicyclo[2.2.2]octane gave a mixture of aldehydes epimeric at C 2 . The aldehyde is then reacted with methyl (triphenylphosphoranylidene)acetate in a Wittig condensation to extend the carboxaldehyde group to a three-carbon, acrylate ester side chain and give compound (III). Catalytic reduction of the acrylate using hydrogen and Raney nickel gives the corresponding propionate ester (IV). Hydrolysis of the ester using formic acid gives the corresponding carboxylic acid which cyclizes spontaneously to give the corresponding lactam (V). The lactam is then reduced to the corresponding cyclic amine (VI) using a hydride reducing agent such as aluminum hydride in an inert solvent such as tetrahydro-furan. The protecting benzyl groups are then removed from VI by standard hydrogenation to give the desired polyol product (VII).

The same general procedure can be used to obtain the corresponding quinolizidine except that the appropriate reactants are used in the Wittig Condensation. That is, the aldehyde II is reacted with methyl 3-(triphenylphosphoranylidene)propionate to give a compound which corresponds to III except that the ester side chain contains four carbon atoms and is a 3-butenoate rather than an acrylate. The butenoate is then subjected to the same series of reactions as described above to give the quinolizidine which corresponds to indolizidine VII.

To obtain the compounds in which R 1 and R 2 are both --OH, the unsaturated ester III or the corresponding butenoate is cis-hydroxylated at the double bond using potassium permanganate or osmium tetroxide to give, after selective hydrogenation to remove the N-phenylmethoxycarbonyl group, the α,β-dihydroxy ester corresponding to IV (or the corresponding β,γ-dihydroxybutanoate). The ester is then reacted as described above to give the hexahydroxy compound corresponding to VII or the corresponding quinolizidine.

The quinolizidines of the present invention in which R 1 is H and R 2 is OA 4 can be obtained from the cyclic carbamate VIII shown in Scheme B below. This Scheme specifically shows the conversion of VIII to a monoester XVII. ##STR4##

Specifically, the cyclic carbamate VIII is treated with aqueous base, such as 50% sodium hydroxide, to open the carbamate and give the hydroxymethyl compuond IX. This alcohol is then reacted with benzoyl chloride to give the N-benzoyl benzoate ester (X). Reduction of the N-benzoyl compound with a hydride reducing agent, such as aluminum hydride, gives the corresponding N-benzyl compound which, after alkaline hydrolysis, gives the N-benzyl hydroxymethyl compound (XI). Swern oxidation of this hydroxymethyl compound gives carboxaldehyde XII. The carboxaldehyde XII can also be epimerized by treatment with hindered bases such as 1,8-diazabicyclo[5.4.0]undec-7-ene or 1,4-diazabicyclo [2.2.2]octane. Reaction of the aldehyde with allylmagnesium chloride gives the carbinol XIII. This reaction gives a mixture of the epimeric carbinols which is carried through the reactions as indicated until the isomers can be separated conveniently. This carbinol is then reacted with an acid chloride to give the corresponding ester. Specifically, the benzoate is shown as compound XIV in Scheme B. Treatment of the unsaturated benzoate with borane followed by hydrogen peroxide gives the corresponding primary alcohol (XV). This procedure also gives some of the 1,4-diol (compound XVIII in Scheme C). The hydroxy ester (XV) is then reacted with cyclohexene in methanol and Pearlman's catalyst to remove the N-benzyl group. The resulting compound is then reacted with methanesulfonyl chloride to bring about cyclization and give the quinolizidine ester XVI. Catalytic hydrogenation using palladium black removes the benzyl protecting groups to give the tetrahydroxy ester XVII.

In a similar fashion, reaction of the carboxaldehyde XII instead with vinyl magnesium chloride gives the allylic alcohol with one less carbon in the side chain than carbinol XIII. The allylic alcohol is then subjected to the same series of reactions as described above to give indolizidine compounds corresponding to quinolizidine compounds XVII (above) and XXI (below). Preparation of the pentahydroxy compound corresponding to XVII is shown in Scheme C below. ##STR5##

Specifically, the 1,4-diol (XVIII), obtained as described earlier, is reacted with cyclohexene in methanol and Pearlman's catalyst to remove the N-benzyl protecting group and give piperidine (XIX). Reaction of the piperidine with methanesulfonyl chloride brings about cyclization to give quinolizidinol (XX). Catalytic hydrogenation of this compound using palladium black removes the benzyl protecting groups to give the desired pentahydroxy compound (XXI).

To obtain the esters of the present invention, besides those whose preparation has been described above, a polyhydroxy compound of the present invention is reacted with an appropriate acid chloride or anhydride. The resulting mixture of esters can be separated chromatographically to give individual monoesters and diesters.

Alternatively, it is possible to use a polyhydroxy compound of the present invention and protect two of the hydroxy groups as ketals and react the resulting protected compound with an appropriate acid chloride or anhydride to esterify a free hydroxy group. The protecting group is then removed to selectively give an ester. Specifically, 5R(5α,6β,7 α,8β,8aα)]octahydro-6,7,8-trihydroxy-5-hydroxymethylindolizine or 6R-(6α,7β,8α,9β,9aα]nonahydro-7,8,9-trihydroxy -6-hydroxymethylquinolizine is reacted with 2methoxypropene or 2,2-dimethoxypropane to give a cyclic ketal between the hydroxy of the more reactive hydroxymethyl group and the hydroxy on the adjacent ring carbon atom. Reaction of this cyclic ketal with an acid chloride such as butyryl chloride or benzoyl chloride preferentially gives the 8-esterified indolizidine or 9-esterified quinolizidine. The ketal protecting group is then removed by treatment with an acid such as HCl in ethanol or 4-toluenesulfonic acid to give the desired monoester.

Alternatively, the initial cyclic ketal obtained above can be reacted with carbobenzoxy chloride to give the same (8- or 9)-benzyloxycarbonate monoesters as described above. These esters can then be further reacted with an acid chloride such as butyryl chloride or benzoyl chloride to give the 7-esterified indolizidine or 8-esterified quinolizidine. The resulting product is then hydrogenated catalytically to removed the benzyloxycarbonate protecting group and then the ketal protecting group is then removed by treatment with an acid such as 4-toluenesulfonic acid to give the monoester(i.e., the indolizidine 7-monoester or quinolizidine 8-monoester).

The present compounds are useful in the treatment of diabetes. More specifically, they can be used to prevent the development of hyperglycemia which may be observed in certain diabetic conditions when a glucose precursor is ingested. Thus, when carbohydrate is ingested either as glucose or in a form such as maltose, sucrose or starch in food or drink, the serum glucose level rises to elevated concentrations. In healthy subjects, this hyperglycemic state quickly returns to normal, the glucose in the blood being rapidly metabolized and stored and/or utilized by the organism. In diabetes mellitus, however, the glucose tolerance of the patient is lowered and the abnormally high serum glucose levels which develop remain elevated for prolonged periods of time. A similar response to that seen in man can also be observed in other animals, including livestock, poultry, pet animals and laboratory animals. Such a condition can be described as postprandial hyperglycemia. One method for treating such a condition would be by administration of some agents which would prevent the conversion of complex sugars to glucose and thus prevent the development of the excessive glucose levels. In the present invention, it has been found that, where the high levels of glucose are a result of the hydrolysis of complex sugars, administration of the present compounds inhibits the initial formation of glucose in the blood and thus makes it possible to avoid the problems which would be associated with prolonged high levels of serum glucose.

The mechanism whereby this result is achieved is the following although the utility described above should not be limited by the precise details of this mechanism. Enzymes which catalyze the hydrolysis of complex carbohydrates convert non-absorbable carbohydrate into absorbable sugars. The rapid action of these enzymes lead to acute and undesirable elevations in blood glucose in diabetes. The compounds of the present invention are potent inhibitors of these enzymes and, when co-administered with a carbohydrate meal, they prevent harmful hyperglycemic excursions of this type. It is desirable, however, that the inhibition of these hydrolytic enzymes be limited to those present in the intestines and that is true for the present compounds. Otherwise, inhibition of systemic glycohydrolases or glucose transport can lead to difficulty in the utilization of intracellular carbohydrates as an energy source and thus cause metabolic problems.

The following test procedure can be used to demonstrate the activity of the present compounds.

In vitro studies. Intestinal glucohydrolases were isolated from rat intestine. Male 150- to 250-g rats were fasted overnight and sacrificed by CO 2 anesthesia. The entire small intestine was removed, flushed with 50 to 100 ml of cold saline and placed on an ice-cold glass plate. The mucosal layer was removed and homogenized in 5 times its volume of 0.5 M NaCl, 0.5 M KCl and 5 mM EDTA, pH 7.0. This homogenate was centrifuged at 20,000× g for 30 min and the pellet washed by suspension and recentrifugation three times in fresh salt solution. The resulting pellet was finally homogenized in 5 times its volume of 0.9% NaCl and centrifuged at 200× g for 10 min. The incubation mixtures contained 10 μl of this enzyme preparation plus 3.3 μmole of sucrose and test compound in a final volume of 100 μl of 0.1 M sodium maleate buffer, pH 5.9. All determinations were performed in duplicate or triplicate. One set of determinations was heat inactivated at 90° C. for 2 min immediately after adding the enzyme. The others were incubated for 30 min at 37° C. in a water bath and then heat inactivated. Glucose concentrations were determined by glucose dehydrogenase (Seragen Diagnostics, Indianapolis, Ind.). Glucose produced at each concentration of test compound was compared with that produced with no drug present.

When [5R-(5α,6β,7α,8β,8aα)]-octahydro-6,7,8-trihydroxy-5-hydroxymethylindolizine was tested according to this procedure, it showed an IC 50 of about 2μM against sucrase.

In practicing a method of this invention, an amount of one of the compounds effective to inhibit postprandial hyperglycemia is administered to a mammal in need thereof by a suitable route. For the purposes of this invention, oral administration is preferred.

The effective amount of the compound, that is, the amount sufficient to inhibit postprandial hyperglycemia, depends on various factors such as the size, type and age of the animal to be treated, the particular compound or pharmaceutically acceptable salt employed, the frequency of administration, the severity of the condition and the time of administration. Generally speaking, the compounds would be administered orally at a dose of 0.5 mpk to 50 mpk, with a dose of 1.5 mpk to 15 mpk being preferred. More specifically, the present compounds would be administered to humans in single unit doses containing 100 mg to 1 g of active ingredient with the material being administered three time a day at mealtime.

The compounds of the present invention are also useful as anti-viral agents and, more particularly, they are useful against retroviruses such as HIV. This utility is established by the fact that the present compounds are inhibitors of α-glucosidase I and thus block the first reaction step in glycoprotein processing of the viral envelope glycoproteins. Without this first biosynthetic step, the virus can not function properly and it would be inhibited. Thus, it has been established that inhibitors of α-glucosidase I would also be useful as antiviral agents and particularly as inhibitors of retroviruses such as HIV. [Sunkara et al. Biochem. Biophys. Res. Commun., 148 (1), 206-210 (1987); Tyms et al. Lancet, ii, 1025-1026 (1987).]The activity of the present compounds as inhibitors of α-glucosidase I can be demonstrated by the following test procedure.

Preparation of [ 3 H] Glucose Labeled Substrate:

The [ 3 H] glucose labeled oligosaccharides substrate (G 3 M 9 N) for glucosidase I was prepared by metabolically labeling exponentially growing BHK cells with [ 3 H]galactose in the presence of 200μg/ml of castanospermine. BHK cells grown as monolayer were treated with 200μg/ml of castanospermine in DMEM (#430-1600) supplemented with 10% heat inactivated fetal calf serum, 2 mM L-glutamine and 1X of PSN antibiotic mixture. After three hour incubation with castanospermine, [1- 3 H] galactose (10μci/ml of media) was added to label the glycoproteins and cells were allowed to grow to confluency for an additional 48 hours. At the end of labeling period, the cells were washed with cold PBS and scraped with a rubber policeman. Cell pellet was heated for 10 minutes at 100° C. and exhaustively treated with pronase (usually 72 hrs) in 50 mM Tris pH 7.5 containing 10 mM CaCl2 and 1% Pronase under toluene atmosphere to obtain glycopeptides. The glycopeptides were separated on columns of Bio-gel P-4. The glycopeptides peak produced by castanospermine treatment was pooled, treated with Endo-H to release the oligosaccharides. The oligosaccharides obtained by endo-H hydrolysis were bound to a ConA column previously washed with buffer A (50 mM Tris pH 7.5 containing 500 mM NaCl) and equilibrated with buffer B (5 mM Sodium acetate buffer pH 5.5 containing 2 mM of each CaCl 2 , MgCl 2 and MnCl 2 ). The oligosaccharides were then eluted with buffer B containing 100 mM α-methylmannoside. The oligosaccharides eluted from ConA column were further purified and characterized on a calibrated Biogel P-4 column (1.5×200 cm, (-) 400 mesh). The purified oligosaccharides having GLc 3 Man 9 GlcNAc structure were used as substrate in these studies.

Preparation of Test Compounds

Compounds were dissolved in H 2 O or DMSO as appropriate and usually 0.02 to 100 μg/ml concentration of the compound was added to the enzyme before starting the reaction with the radioactive substrate. DMSO controls were run for each experiment, if DMSO was used to dissolve the compounds.

Microtiter Plate Assay for α-Glucosidase I Activity

ConA-sepharose was washed first with buffer A, then with buffer B as described above, and resuspended in buffer B (gel: buffer, 1:1) before use. The enzymatic assays were performed in a 96 well microplate in a total volume of 100 μl which contained 5000 CPM of [ 3 H]G 3 M 9 N substrate, 100 mM potassium phosphate buffer pH 6.8 and purified α-glucosidase I. The reaction mixture was incubated at 37° C. for one hour for each experiment and the reaction stopped by adding 25 μl of glacial acetic acid. To the mixture, 175 μl of concanavalin A-sepharose in buffer B (1:1) was added and microplate was spun at 500× g for 5 minutes. A 150 μl aliquot of supernatant was removed and counted. When Compound VII in Scheme A above was tested by this procedure, it showed an IC 50 of 0.3 μM. When compound XVII in Scheme B above was tested by this procedure, it showed an IC 50 of 29μM. Compound XXI in Scheme C above showed an IC 50 of 0.15μM in this test.

Inhibition of Glucosidase I in F-10 Cells:

Accumulation of G 3 (G 3 MsN 2 -Asn) in F-10 cells is used as a measure of inhibition of α-91ucosidase I. The qlycopeptides obtained (as above for BHK cells) by pronase digestion of radio labeled F-10 cells in the presence of different concentration of inhibitors (usually 0.1 to 30 μg/ml) are chromatographed on Bio-gel PD-6. The relative percentage of counts in G 3 peak as compared to void volume (where the radioactivity in the controls is eluted) are used as a measure of inhibitory activity of a particular compound.

The compounds of this invention can be used to treat a number of diseases and conditions known to be caused by pathogenic viruses including those diseases and conditions caused by murine leukemia virus, feline leukemia virus, cytomegalo-virus (CMV), avian sarcoma virus, human immunodeficiency virus (HIV), HTLV-I, and HTLV-II. Those experienced in this field are readily aware of the circumstances requiring anti-retroviral therapy. Applicants consider the use of the compounds of this invention to treat HIV infections in humans to be of most importance. The term "patient" used herein is taken to mean mammals such as primates, including humans, sheep, horses, cattle, pigs, dogs, cats, rats and mice.

The amount of a compound of the present invention to be administered can vary widely according to the particular dosage unit employed, the period of treatment, the age and sex of the patient treated, the nature and extent of the disorder treated, and the particular parent compounds or ester derivative selected. Moreover the derivative can be used in conjunction with other agents (e.g. AZT) known to be useful in the treatment of retroviral diseases and agents known to be useful to treat the symptoms of and complications associated with diseases and conditions caused by retroviruses. The anti-retrovirally effective amount of a compound of the present invention to be administered will generally range from about 15 mg/kg to 500 mg/kg. A unit dosage may contain from 25 to 500 mg of the parent compound or ester derivative, and can be taken one or more times per day. The compound used can be administered with a pharmaceutical carrier using conventional dosage unit forms either orally or parenterally.

In practicing the methods of this invention, the active ingredient is preferably incorporated in a composition comprising a pharmaceutical carrier and from about 5 to about 90 percent by weight of a compound of the invention or a pharmaceutically-acceptable salt thereof. The term "pharmaceutical carrier" refers to known pharmaceutical excipients useful in formulating pharmaceutically active compounds for internal administration to animals, and which are substantially non-toxic and non-sensitizing under conditions of use. The compositions can be prepared by known techniques for the preparation of tablets, capsules, elixirs, syrups, emulsions, dispersions and wettable and effervescent powders, and can contain suitable excipients know to be useful in the preparation of the particular type of composition desired. Suitable pharmaceutical carriers and formulation techniques are found in standard texts, such as Remington's Pharmaceutical Sciences, Mack Publishing Company, Easton, Penna.

The following examples are presented to illustrate the present invention. However, they should not be construed as limiting it in any way.